AI Paper Atlas

Research map / Label-efficient and robust learning / Domain adaptation and generalization

Moment Matching for Multi-Source Domain Adaptation

Xingchao Peng, Qinxun Bai, Xide Xia, Zijun Huang, Kate Saenko, Bo Wang

ICCV 2019 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Feature alignment for unsupervised adaptation · 1,822 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add source-source moment alignment to source-target alignment (ablation)M3SDA-beta · Digit-Fiveaccuracy (%): 85.8 -> 87.7Improved
add source-source moment alignment to source-target alignment (ablation)M3SDA-beta · Office-Caltech10accuracy (%): 96.2 -> 96.4Improved
add source-source moment alignment to source-target alignment (ablation)M3SDA-beta · DomainNetaccuracy (%): 39.7 -> 42.6Improved
add source-target moment alignment to source-source alignment (ablation)M3SDA-beta · DomainNetaccuracy (%): 34.4 -> 42.6Improved
add source-target moment alignment to source-source alignment (ablation)M3SDA-beta · Digit-Fiveaccuracy (%): 81.5 -> 87.7Improved
add dynamic weight vector over source classifiers (ablation)M3SDA · DomainNetaccuracy (%): 40.8 -> 41.5Improved
add classifier-discrepancy training (beta variant) (ablation)M3SDA · DomainNetaccuracy (%): 41.5 -> 42.6Improved
add classifier-discrepancy training (beta variant) (ablation)M3SDA · Digit-Fiveaccuracy (%): 86.13 -> 87.65Improved
add classifier-discrepancy training (beta variant) (ablation)M3SDA · Office-Caltech10accuracy (%): 96.1 -> 96.4Improved
combine multiple sources into one domain for single-source DANDAN · DomainNetaccuracy (%): 32.9 -> 32.1Hurt
multi-source adaptation with quickdraw targetM3SDA · DomainNet quickdraw targetaccuracy (%): 13.3 -> 5.2Hurt
increase number of categories from 20 to 345state-of-the-art DA methods · DomainNet painting->realaccuracy (%)Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy